Hello fellow UiPath professionals, I am looking for options and suggestions for one of the process requirements I am working on. We are looking to use an input application file with input data like name , dob to find that person in another system for existence . Now the person entering data in the application does some typos or enters data in different format like first name , last name making it difficult to find that person in the second system even though the record actually exists . We have tried using the fuzzy component from the marketplace but isn’t helping catch certain scenarios. Appreciate your inputs
Welcome back to the forums @rinki
I think the input file needs to be fixed so you are forcing the user to provide clear data.
I would suggest updating the input file to have separate columns for First Name, Middle Name, Last Name and DOB
Then you can stitch together the name in the format best for searching the application.
Hopefully this helps,
Cheers
Steve
Can you give me a couple of samples? Name in system A and Name in System B.
I recently worked in a library that combines Levenshtein distance algorithm in a token based approach (comparing word by word and calculating an overall score). It is a work in progress, so if you can share this sample data, I can share the results with you in terms of similarity scores that this activity provides.
Hi @rinki
I would agree with @Steven_McKeering on this.
First system where user is entering not in desired format, it would be good to have validations in place to verify the data being inputted and even throwing descriptive errors to user to input the data in right format.
If that is something you guys are not looking to incorporate at the moment, you can consider doing a review of the sheet being generated from first system that would act as an input for second system. If any errors found in the sheet, have it corrected before using it in the 2nd application.
This might be a bit of manual work.
Hope this helps.
Regards
Sonali